Add BOOTSNAP_READONLY to Dockerfile template for production

[technicalpickles]

Motivation / Background

This was added in back in bootsnap 1.15.0 (Shopify/bootsnap#428). This can save extra work and IO when the cache won't be changed, like is the case for docker containers

Detail

This Pull Request changes the Dockerfile template used when creating a new application to add BOOTSNAP_READONLY=1

Additional information

Checklist

Before submitting the PR make sure the following are checked:

• This Pull Request is related to one change. Changes that are unrelated should be opened in separate PRs.
• Commit message has a detailed description of what changed and why. If this PR fixes a related issue include it in the commit message. Ex: [Fix #issue-number]
• Tests are added or updated if you fix a bug or add a feature.
• CHANGELOG files are updated for the changed libraries if there is a behavior change or additional feature. Minor bug fixes and documentation changes should not be included.

[byroot]

This shouldn't make much of a difference in this context, Docker preserve mtime so bootsnap shouldn't have anything to update.

But actually, would need to measure, but I suspect this is actually harmful because you set it before rake assets:precompile, so I suspect Bootsnap won't be able to persist its load path cache, marking the final image boot slower.

[technicalpickles]

Someone pointed out that the ENV would need to go into final image 😅 I think that would mitigate the concern about precompile since it wouldn't be read-only until the image is run.

@byroot do you have any thoughts on how to measure the impact on this? Time to first request when running the docker image?

[byroot]

do you have any thoughts on how to measure the impact on this?

Yeah, just need time bin/rails runner 1 in production mode.

You can also set BOOTSNAP_LOG=1 to see if there is cache misses.

[byroot]

I've tested it and it drastically changes application boot time: down from near 2mn to 30s 😃

Edit: the app runs in Docker, on HDD drives

That's in "too good to be true" territory. Can you try just disabling Bootsnap to compare?